Field of the Invention
The present invention relates to a recording medium, an information processing apparatus, and a depth definition method and, more particularly, to a technique of defining an undulation in a depth direction for a 2D image.
Background Art
In recent years, in the technical field of computer graphics represented by an electronic game, rendering representation using a 3D model has become the mainstream. This is because when, for example, rendering frames at a plurality of successive points while moving the same character and the background like animation, the time taken to perform processing such as repetitive rendering processing or lighting calculation processing can be reduced. Especially for an interactive content in which the action of a character and the like change in real time in response to an operation input in an electronic game or the like, it is possible to cope with rendering representations from various viewpoint positions or directions by preparing in advance a 3D model, animation data, and the like. In general, in such 3D graphics, a 3D model is constructed based on a plurality of 2D images (cuts) prepared by a designer or the like, and texture is applied to the model, thereby performing rendering.
On the other hand, in 3D graphics rendered by applying texture, an impression different from the initial cut prepared by the designer or the like may be given. The 3D graphics is basically a technique for “correctly” rendering a 3D model applied with texture with respect to a specific viewpoint, and it is thus difficult to reproduce a representation of an effective “appearance” in a specific line-of-sight direction, unlike a cut rendered by a 2D image. Therefore, a game content in which a 2D image is mainly used in a game screen by prioritizing the appeal of the representation unique to the 2D image has been supported to some extent.
PTL 1 discloses a rendering technique capable of representing (three-dimensionally representing) 3D animation while maintaining the atmosphere and appeal of a 2D image rendered by a designer or the like. More specifically, in PTL 1, after dividing a 2D image into parts such as hair, eyebrows, eyes, and a contour (face), a standard curved surface is simply assigned to the part of the contour in accordance with the appearance of the 2D image. By performing geometric deformation and movement for the 2D images of other parts in accordance with the spherical surface of the contour rotated in correspondence with the direction of the face to be represented, and allowing application of various adjustment operations, a representation desired by the designer from a different direction is implemented without damaging the impression of the original 2D image. That is, unlike a method of performing rendering by simply applying texture, the method described in PTL 1 adopts a method of deforming a 2D image to implement a representation desired by the designer.